Psychological inflexibility and resilience in anxiety: insights from machine-learning and robust mediation-based models
Frontiers in Psychiatry · 18.05.2026 · 17
DOI: 10.3389/fpsyt.2026.1769001 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Psychiatry, online erschienen am 18.05.2026. Verfasst von Ana Ferreira Mostardinha und 6 weiteren Autor:innen. Beteiligte Einrichtung: Coimbra Institute for Biomedical Imaging and Translational Research (CIBIT), Institute for Nuclear Sciences Applied to Health (ICNAS), University of …. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Psychological inflexibility (PI) has been associated with anxiety symptoms, while resilience serves as a protective factor; however, their roles and interrelationship remain poorly understood. We investigated the role of PI on anxiety-related symptoms while assessing the mediating role of resilience and testing the moderating effect of sex and psychiatric history.
Methods
From April to July 2021, an online protocol employing self-reported measures assessed PI (Acceptance and Action Questionnaire), resilience dimensions (Resilience Scale for Adults), and anxiety-related symptoms (Generalized Anxiety Disorder (GAD) Scale; Depression, Anxiety, and Stress Scales). A model generation approach, using machine-learning and robust mediation-based models, was applied to investigate the relationships between these constructs.
Results
In a sample of 313 adults (72.20% females; 39.29 ± 11.81 years), Random Forest analysis indicated PI and the resilience dimensions perception of self (R-PS) and planned future (R-PF) as the strongest predictors of anxiety-related symptoms. PI showed a positive direct association with GAD, anxiety, and stress (respectively β = 0.28, β = 0.07, β = 0.20, p ≤ 0.001). Significant indirect associations emerged: PI–Stress regarding R-PS (β = 0.08, p = 0.004), PI–Anxiety regarding R-PF (β = 0.03; p = 0.03), PI–GAD (β = 0.08, p = 0.001) and PI–Stress (β = 0.11, p < 0.001) regarding R-PS and R-PF together.
Discussion
These findings highlight the importance of PI and resilience as interconnected processes underlying mental health outcomes. Additionally, they suggest that psychological intervention programs targeting PI, along with resilience, could foster healthier strategies for coping with anxiety-related symptoms.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Coimbra Institute for Biomedical Imaging and Translational Research (CIBIT), Institute for Nuclear Sciences Applied to Health (ICNAS), University of Coimbra, Coimbra, Portugal.
- Faculty of Medicine (FMUC), University of Coimbra, Coimbra, Portugal.
- Instituto Superior Miguel Torga (ISMT), Coimbra, Portugal.
- Institute of Electronics and Informatics Engineering of Aveiro (IEETA), University of Aveiro, Aveiro, Portugal.
- School of Health and Social Care, University of Essex, Colchester, United Kingdom.
- Centre for Informatics and Systems (CISUC), Department of Informatics Engineering (DEI), University of Coimbra, Coimbra, Portugal.
Zitieren
APA
Mostardinha, A. F., Sousa, D., Senra, H., Simões, M., Pina Rodrigues, A., Vagos, P., & Castelo-Branco, M. (2026). Psychological inflexibility and resilience in anxiety: insights from machine-learning and robust mediation-based models. Frontiers in Psychiatry, 17. https://doi.org/10.3389/fpsyt.2026.1769001
Vancouver
Mostardinha AF, Sousa D, Senra H, Simões M, Pina Rodrigues A, Vagos P, et al. Psychological inflexibility and resilience in anxiety: insights from machine-learning and robust mediation-based models. Frontiers in Psychiatry. 2026;17. doi:10.3389/fpsyt.2026.1769001
BibTeX
@article{Mostardinha2026,
author = {Mostardinha, Ana Ferreira and Sousa, Daniela and Senra, Hugo and Simões, Marco and Pina Rodrigues, Ana and Vagos, Paula and Castelo-Branco, Miguel},
title = {Psychological inflexibility and resilience in anxiety: insights from machine-learning and robust mediation-based models},
journal = {Frontiers in Psychiatry},
year = {2026},
volume = {17},
doi = {10.3389/fpsyt.2026.1769001}
}
EUVIMED